Cyber Risks · The Hacker News

LiteLLM CVE-2026-42208 SQL Injection Exploited within 36 Hours of Disclosure

Summary: A critical SQL injection in BerriAI's LiteLLM Python package, tracked as CVE-2026-42208 with a CVSS score of 9.3, was exploited in the wild within 36 hours of disclosure, with the first attempt recorded on April 26 at 16:17 UTC, roughly 26 hours and seven minutes after the GitHub advisory was indexed. The flaw let an unauthenticated attacker send a crafted Authorization header to any LLM API route and reach a proxy API key check query that mixed caller-supplied input into the query text, and Sysdig observed the activity from IP 65.111.27[.]132 targeting tables such as litellm_credentials.credential_values and litellm_config, followed about 20 minutes later by a second probe from 65.111.25[.]67.

Why it matters: Sysdig notes that a single litellm_credentials row often holds an OpenAI organization key with five-figure monthly spend caps, an Anthropic console key with workspace admin rights, and an AWS Bedrock IAM credential, so a successful database extraction is closer to a cloud-account compromise than a typical web-app SQL injection. CISA added the flaw to its KEV catalog on May 8, 2026, with a patch deadline of May 11, 2026 for federal civilian agencies.

Read the original at The Hacker News ↗

Topics
Vulnerabilities & Patching
Mentions
CVE-2026-42208
Related
CVE-2026-42208: Berri AI LiteLLM SQL Injection Vulnerability Known Exploited Vulnerabilities · May 08, 2026